19 Mar
2011
There's been a lot of talk recently in the web technology sector of a new bubble. Valuations are increasing across the board, the demand for talent is growing, and everyone and their mother appears to be working on a startup. All this bubble talk has got me thinking about what really constitutes a bubble and if its actually possible to recognize a bubble while its actually happening.
The best rule that I've come up with to gauge this is the presence of what I like to call "multicollinear expectations." Looking at documentation of the real estate bubble a few years ago and talking to people about the education sector (which I suspect is also in the midst of a bubble) I realized that the key to understanding whether or not a sector/industry is in a bubble is to see if people's expectations of the future of that sector are positively affected by other people's optimism. For example, if you were to take five people and individually ask them their opinion of the future of a sector they would likely give you their honest opinions (and their predictions would probably be quite reasonable). However, if you place them in a room together and ask the group for their expectations of the future of a sector, if a bubble were occurring you'd see that the average prediction for the group together was actually higher than the average of each individual's opinion.
In most bubbles, this type of behavior tends to manifest itself as a fear of being left out or left behind. This feeling, when our expectations of the future become overly dependent upon each other's perceptions of the future and result in a collective view that is more optimistic than each of our independent views, is arguably what leads to bubble-like behavior. I believe that this can be used as a pretty good indicator for judging if a bubble exists.
So, the $64,000 question: Does this hold true for the web/mobile/consumer technology sector today?